Working for Cross-Border Indicators

Working for C02 Emissions Embodied in Final Domestic Demand

Hence the values are not available for each country for year 2016-2019

selction of relevent columns

3 Models are used for the purpose

Linear Regression

Random Forest Regression

MLP Regression

The above analysis show that MLP Regression has reletively closer results to the Actual CO2 when it comes accuracy and error calculations

USING MLP to predict 2016,2017,2018 and 2019

Working for C02 Emissions Embodied in Gross Exports

C02 Emissions Embodied in Gross Exports is not available for 9 countries

Working for not available data

Linear Regression

Random Forest Regression

MLP

the above figure shows that in most of the cases MLP is better then RF and LR

MLP to predict missing data

Working for C02 Emissions Embodied in Gross Imports

Linear Regression

Random Forest

MLP

the above analysis shows that MLP is better then other 2 in most cases

Working for C02 Emissions Embodied in Production

Linear Regression

Random Forest Regression

MLP

Combined

MLP and LR has better accurcies but MLP is slightly less error prone then LR

Major Exporters and importersof CO2

C02 Emissions Embodied in Final Domestic Demand relations with Exports/imports

C02 Emissions Embodied in Production relations with Exports/imports